xvi] VIBRIO AND SPIRILLUM 463 



cholera that had occurred in Massowah. Pfeiffer accepted 

 this Massowah vibrio as the cholera vibrio — notwithstanding 

 its slight deviations in cultural respects from the typical 

 Koch's vibrio, no doubt influenced by the knowledge gained 

 that the vibrios derived from undoubted cases of cholera do 

 not all coincide in all cultural characters— and his earlier 

 experiments and statements on cholera were admittedly 

 made with, and refer to this Massowah vibrio. Metchnikoff 

 also accepted, after study, the Massowah vibrio as the true 

 cholera vibrio ; many of his experiments and observations 

 on animals and human beings were made with this vibrio. 

 Now, unfortunately this Massowah vibrio does not give 

 Pfeiffer's test, and therefore is declared by Pfeiffer not to be 

 cholera vibrio at all. This is a difficulty, though like all 

 such difficulties it need not deter us from altering an initial 

 wrong conclusion. But there are other and greater diffi- 

 culties. Pfeiffer cannot deny the possibihty that vibrios 

 originally derived from true cholera, but living afterwards 

 under abnormal conditions of temperature, soil and others, for 

 considerable periods, could so alter as to change some of their 

 original cultural characters as also their physiological reactions. 

 This, for instance, seems to me to have been the case 

 with Sanarelli's water vibrios, with Pestana's vibrio, and I 

 have already given direct evidence of such being the case 

 with my oyster vibrios. There is nothing extraordinary or 

 new in such an assumption ; it is borne out by laboratory 

 observations on a number of microbes, altering their charac- 

 ters permanently, cultural and chemical, by the influence of 

 medium, temperature, the animal body, &c. One could 

 therefore well assume or at any rate admit the possibility — 

 it would be no exaggeration even to say the probability — 

 that cholera vibrios living in water might or would so alter 

 that the nature of their behaviour under Pfeiffer's test might 



